Abstract: Milk yield from adapted cows is facing problems due to varying climatic conditions and adaptability of imported animals. The main objective of study was to describe the relationships between body condition scores and daily milk yield across different days in milk. Genetic (co) variances between Body Condition Score (BCS) and test-day milk yield (TDMkkg) were estimated using a random regression animal model extended to multivariate analysis. A cubic random regression was sufficient to model the changing genetic variances for BCS and milk across different Days in Milk (DIM). The milk yield data was obtained from the electronic herd records. The animal groups (milking cows) were evaluated according to the milk yield level and body condition scores. The additive genetic correlations between BCS and milk production showed high change over the lactation duration (pooled or separate parties). The highest additive correlations between BCS and TDMkkg were around 0.6 across 80-125 days in milk. Body condition scores during early and late part of lactation were negative and low related to milk production. This suggests that the highest genetic improvement in daily production from indirect selection on BCS should be based on measurements taken during the mid-lactation stage when the additive genetic variance for BCS is the largest.